High Speed Downlink Packet Access (HSDPA) is a feature that was introduced in Release 5 of the third generation partnership project (3GPP) specifications to operate in a cell dedicated channel (Cell_DCH) state. A key operating principle of HSDPA is to share a very fast downlink (DL) pipe, the high-speed downlink physical shared channel (HS-DPSCH). A universal terrestrial radio access network (UTRAN) can configure up to 15 HS-DPSCHs, and each of these can by shared by all wireless transmit/receive units (WTRUs) on a per transmission time interval (TTI) basis, (i.e., every 2 msec). As a result, the information on the downlink channels can be destined to different WTRUs at an interval of every 2 msec. In order to allow WTRUs to determine ownership of the information on these shared channels, a Node-B also sends a set of parallel high speed shared control channels (HS-SCCHs). Among other things, these channels provide details allowing the WTRUs to determine if the information on the HS-DPSCH is for them, and if so, to recover the transmitted information.
In HSDPA, the Node-B tries to make a better use of the downlink capacity by using several concepts, such as adaptive modulation and coding (AMC), retransmissions using a hybrid automatic repeat request (H-ARQ) scheme, and Node-B scheduling. All of these operate at a very fast rate, directed toward taking advantage of the changing channel conditions perceived by the WTRUs. In order to accomplish this, the Node-B schedules transmissions in order to maximize downlink throughput, (e.g., using 16-quadrature amplitude modulation (QAM) for a WTRU very close to the Node-B and using quadrature phase shift keying (QPSK) for a WTRU at cell edge). This fast scheduling is complemented with H-ARQ at a physical layer, allowing retransmission of transport blocks that are received in error. Furthermore, multiple simultaneous H-ARQ processes are permitted in order to maximize utilization.
Every 2 msec, the Node-B schedules transmissions on the HS-DPSCH based on WTRU channel conditions, the status of its own downlink buffers, and status of the parallel H-ARQ processes. In addition, the Node-B tailors the transmission bit rates to the specific WTRUs by adapting the modulation, coding, and transport block size. To achieve this, the Node-B requires the following information from the WTRUs:
Channel Quality Indication (CQI): an index into a table which provides the maximum MCS and transport block the WTRU can support, based on the received DL signal power, the interference from other cells, and the WTRU receiver capability; and
Positive acknowledgement (ACK)/negative acknowledgement (NACK) feedback for the H-ARQ processes.
This feedback information is provided on a high speed dedicated physical control channel (HS-DPCCH). The CQI information is provided periodically, with periodicity determined by the UTRAN. The ACK/NACK information is only provided in response to a packet being received on the downlink. Some of the main attributes of the HS-DPCCH include:
Every WTRU having an access to the HSPDA has its own dedicated control channel and as a result the WTRU can easily provide the feedback information;
The channel is power controlled; this is accomplished through a closed loop mechanism using the downlink dedicated physical control channel (DPCCH), which always accompanies HSDPA transmission;
The information on the HS-DPCCH is heavily coded to aid in detection; and
The information on the HS-DPCCH is time aligned (but delayed) with respect to the UL DPCCH.
As part of Release 7 of 3GPP, using the high speed shared channel for WTRUs in a Cell_FACH state has been considered. The properties of a WTRU in a Cell_FACH state are quite different from those for a WTRU in a Cell_DCH state including the following:
The only uplink mechanism for Release 7 WTRUs in a Cell_FACH state is through a random access mechanism, over the random access channel (RACH);
The WTRUs autonomously update the UTRAN as to their cell location through CELL UPDATE and URA UPDATE procedures;
The WTRU can be configured for forward access channel (FACH) measurement occasions—these are periods during which the WTRU takes measurements on inter-radio access technology (RAT) and/or inter-frequency cells;
The WTRUs make the following physical layer measurements:                common pilot channel (CPICH) received signal code power (RSCP): The received signal code power measured on the primary CPICH; and        CPICH Ec/No: The received energy per chip divided by the power density in the band. The CPICH Ec/No is defined as CPICH RSCP/UTRA Carrier received signal strength indicator (RSSI). The UTRA Carrier RSSI is not measured in Cell_FACH, but only in Cell_DCH. It is the received wide band power, including thermal noise and noise generated in the receiver, within the bandwidth defined by the receiver pulse shaping filter; and        
Physical layer measurements are reported to the UTRAN via layer 3 signalling through the Measured Results on RACH mechanism.
The measured results on RACH mechanism are designed to provide some feedback information to the UTRAN regarding the current state of the downlink. The mechanism may be controlled by the network, which broadcasts, as part of its system information in system information block 11 (SIB11) and SIB12, the information element (IE) “Intra-frequency reporting quantity for RACH Reporting”. This IE tells all WTRUs in the cell the quantity to measure. The choices are highlighted in Table 1 below, along with the granularity of the reported quantity and the length of the coded information sent to the UTRAN.
TABLE 1Size of CodedQuantityDetermined by/fromGranularityInformationCPICHas measured by PHY layer−120 to −25 dBm7RSCP(in steps 1 dBm)CPICHas measured by PHY layer−24 dB to 0 dB6Ec/No(in steps of 0.5dB)PathlossPathloss in dB =46 to 158 dB7Primary CPICH(in steps of 1Tx power-CPICH RSCP.dB)Where Primary CPICH Tx power isfound in the IE “Primary CPICH Txpower”, and CPICH RSCP is asmeasured in PHY layerNo0report
Referring to Table 1 above, a WTRU takes the appropriate measurements and reports these to the higher layers, (e.g., radio resource control (RRC)). In a Cell_FACH state, the base measurement period is 200 msec, but this can be higher if a FACH Measurement Occasion is configured for monitoring inter-frequency and inter-RAT cells. The accuracy of the measurement is defined to be ±6 dB for the current cell.
The information is sent to higher layers to be included in the IE “Measured results on RACH”, in a number of RRC messages including: CELL UPDATE, RRC CONNECTION REQUEST, INITIAL DIRECT TRANSFER, UPLINK DIRECT TRANSFER, and MEASUREMENT REPORT messages. Unlike the majority of other measurements and quantities available at the RRC, those carried in the IE “Measured results on RACH” can not be filtered by the RRC prior to transmission.
The IE contains the configured measured quantity for the current cell and all other intra-frequency cells in the monitored set (the monitored set includes a list of cells that the UTRAN has told the WTRU to monitor). The IE may also contain the configured measured quantity for inter-frequency cells.
For RRC messages transmitted on a common control channel (CCCH), the maximum allowed message size should not be exceeded when forming the IE “Measured results on RACH”. To achieve this, the WTRU may limit the number of included neighbor cells included in the report, or omit the IE “Measured results on RACH” altogether.
If HSDPA is to be used in a Cell_FACH state, the main problem is the lack of a dedicated uplink channel and the resulting absence of feedback. Without this information, the advantages of HSDPA are significantly reduced.
Several proposals addressed to solve the feedback issue have already been made. In one proposal, it has been proposed to transmit CQI information over:                The RACH transport channel either in preamble or message;        A new shared common uplink channels; and        Shared uplink code space.        
However, the techniques described above require changes to the WTRU physical layer. As there is a strong push to limit these changes, it has been proposed to use the “Measured Results on RACH” mechanism to transmit channel quality information to the UTRAN. As the information is embedded in an RRC message IE, the UTRAN radio network controller (RNC) would need to forward the channel quality information to the Node-B so that it may perform its adaptive modulation and coding.
Using the “Measured Results on RACH” mechanism to provide channel quality feedback has a number of problems, however. For example:
1. The measured quantity (metric) that is fedback and subsequently used by the Node-B is either CPICH RSCP or CPICH Ec/No. Problems that may be associated with both these measures include:
1) RSCP only measures received power on a single code and does not provide any indication about the neighbour cell interference experienced by the WTRU.
2) Ec/No provides a ratio of the power received on the CPICH to the total received downlink power (CPICH Ec/No=CPICH RSCP/UTRA Carrier RSSI). As the denominator is the total received DL power, the range of the metric is somewhat reduced compared to a pure carrier/interference ratio. Additionally, the downlink received power (i.e., RSSI) is not necessarily measured in the Cell_FACH state. If it is not, then when making a CPICH Ec/No calculation, the physical layer uses the last calculated RSSI value while in Cell_DCH state. If a UE stays in Cell_FACH for an extended period of time, this value will be outdated, and the measure will be no better than the CPICH RSCP.
3) These measures fail to take into account the WTRU receiver capability. It is possible that for the same CPICH RSCP, WTRUs with different receiver architectures could support very different modulation, coding, and transport block sizes.
4) There is poor correlation between the RSCP and Ec/No and the actual CQI.
2. As WTRUs in Cell_PCH move across cell boundaries, the UTRAN refreshes their channel quality information based on information received in the CELL UPDATE message. If these WTRUs transition to Cell_FACH, the UTRAN will have up-to-date channel quality information which it can relay to the Node-B. On the other hand, as WTRUs in URA_PCH cross UTRAN routing area (URA) boundaries, they inform the network with a URA UPDATE message. As this message does not carry the IE “Measured Results on RACH”, the network will typically have out-of-date channel quality information.
3. The quantities reported in the IE: “Measured Results on RACH” cannot be filtered at the WTRU RRC. As there is a latency between the instant the measurement is made and the time the Node-B is privy to this information, using non-averaged results may result in sub-optimum performance.
4. The Measured Results on RACH mechanism is currently controlled on a per-cell basis through information broadcast in the system information. As a result, all WTRUs in a cell are required to report the same information. This limits the flexibility of the reporting mechanism as the UTRAN may favor one measure over the other depending on the eventual use. For instance, the pathloss may be a better measure for cell reselection, but the CPICH Ec/No may be better for channel quality reporting.